[RFC] C++ vs forced unwinding

Benjamin Kosnik bkoz@redhat.com
Wed Apr 2 17:25:00 GMT 2003


>I've spent a *lot* of time thinking about this.  I can't think of
>any solution except to compile with (at least) -funwind-tables.

You've obviously spent more time thinking about this than I have in my
replies... I'm not trying to give you a hard time.

>> It was really nice when this became something that
>> "just worked," and it would be a shame to lose this.
>
>Does it help to know that ia64 and x86_64 *always* include 
>unwind tables, and so it will continue to "just work" there?  ;-)

If I had one of these machines, maybe.

>Really, I think it's about time to revisit re-enabling unwind
>tables by default in C.  The overhead when using current 
>binutils and glibc has dropped to just about zero.

Now this sounds like the real solution for lazy people like me! 

Then, unless I am once again mistaken, everything (meaning C compatible
shared libraries for C++ et. al) would work by default, and the people
who wish to make things difficult for C++ et al could do their own thing
with their own flags. That sounds like the way to go, from my vantage
point.

Plus, this means that all arches would then be doing this in a
consistent way, right? Seems like a win on this alone....

-benjamin



More information about the Gcc-patches mailing list